Telecommunications demand – a review of forecasting · PDF fileTelecommunications demand...

© Robert Fildes, IIF & Lancaster Centre for Forecasting

Telecommunications demand – a review of forecasting issues

Robert Fildes, Lancaster [email protected]

This paper is based on on work published in the IJF: 2002(4) as part of a special issue on Telecoms Demand Forecasting.

The text downloadable from Science Direct


(c) Cable

(b) mobile

(d) satellite

Consumer End Uses•video• games• medical•home shoppingvia• internet• call centres

IXC2 IXC2

IXC POT(a) PSDN IXC

(e) local bypassBusiness End Uses• data transfer• video conferencing• group work• CAD/CAM

Localcarrier


•A Hierarchy of Forecasts

lOverall demand for an application

lCompetition within technologies to service application

lCompetition between suppliers

l Once investment in place ä operating costs are minimal

Forecasting Implicationsä e.g. for entrants, price collusion, price leadership

Usage

Access

Equipment


•

n Stable Markets - aggregateä Public Switched

n Stable Markets - disaggregateä Calling patterns

nGrowth Marketsä Special Servicesä New services

l ADSL

n Operations

Types of Telecommunications Forecasting Problems

Ø Corporate/ regulator data available

Ø Surveys, customer behaviour

Ø Intentions SurveysØ Feature evaluationsØ Trial markets Ø Analogies

– other products– other countries

Ø Delphi/ Expert judgement

Ø Statistical time series methods

Different problems - different organisational responsibilitiesDifferent problems - different organisational responsibilities


Stable Markets

• Point-to-point PSTN– Local, long-distance, international

• Drivers– Price, income, advertising, ‘value-for-money’

• Problem-specific drivers– Call-back on international

• Price elasticity declining– Depends on price, culture

Econometric models withFocus on price elasticity estimation

US international message telephone service (IMTS): (Econometric comparisons, Madden et al, IJF)


Conclusions - Stable MarketsConclusions - Stable Markets

n Increasing econometric sophisticationØ model specification (externalities: market

size)Ø diagnostics Ø multicollinearity - the interrelationship

between the explanatory variables, e.g ownership of durables and income

•But no stability or forecasting tests

•Time varying coefficients?

Missing out or including inappropriate variablesê•mis-understanding the market• poor forecasts,•bad policy

Does the model work better than alternatives?

Is the market changing,are new services getting more acceptable, is quality becoming more important?

Missing out or including inappropriate variablesê•mis-understanding the market• poor forecasts,•bad policy

Does the model work better than alternatives?

Is the market changing,are new services getting more acceptable, is quality becoming more important?


Stable Markets – Disaggregate Stable Markets – Disaggregate

Access & Usage• Based on Cross-sectional household data• Conditional models of usage, given choice of technology

or carrier• Focus on price elasticity

– By household demographics to support universal service analysis– Linked to marketing planning

Evaluation:Despite their use for forecasting• Dynamics of change not taken into account• No forecasting evaluation

Evaluation:Despite their use for forecasting• Dynamics of change not taken into account• No forecasting evaluation


New Products & ServicesNew Products & Services

• Market Potential– mobile– ‘really new products’, e.g. 3G

• Diffusion Path– E.g mobile

• But now market dominated by ‘switchers’?

ØIntentions SurveysØFeature evaluationsØChoice modelsØTrial markets ØAnalogies

– other products– other countries


First generationUp to T

Both generations,Competition

Time line

Adopters at T-1

Market Potential forfirst generation, M1

Adoptersat T

Adopters &Users at T

Leavers Switchers

Exit

Adopters &Users at T+1

Adopters &Users at T+1

Switchers

ExitExit

Market Potential forsecond generation, M2

M1, not M2

Leavers

Forecasting telecommunication service subscribers in substitutive and competitive environments, Jun et al, IJF, 2002


Competition between TechnologiesCompetition between Technologies

0

20

40

60

80

100

120

140

160

0 10 20 30 40 50

Period

1st

Gen

'ion

1st Gen'ion

2nd Gen'ion

TOTAL


Segmentation Approaches to Forecasting)

Small

LargeCommercial

LargeService

Business

Income > 30K

Income < 30K

Professional

BusinessSubscribers

Attending College Not attending

Below 21 years

Income < 20K

21 - 50

Income > 30K Income <30K

Not retired Retired

> 50

PersonalSubscribers

All users

• Forecast consumption in each segment• Project numbers in each segment }and multiply

Market Potential -Decomposition Methods I (applies to new and established markets

Market Potential -Decomposition Methods I (applies to new and established markets


Decomposition Methods IIDecomposition Methods II

Step: 1. Identify the most important services (in terms of load in bits per

second on the network).2. Segment the population, e.g business and consumer, high and low

intensity users.3. Using ideas of a time budget (the number of hours a day available

for telecoms related use) estimate the services each segment might use, by service, and the level of usage.

4. Model the changes in the segments over time.5. Obtain total traffic by aggregating the individual forecasts of the

usage profiles.


Alternative Models of choice& Usage - for households with Internet Access

(see Rappaport et al, 2001)

Simultaneous Sequential

?

PSDN ADSL ISDN2

?

PSDN Broadband?

ADSL ISDN2

Usage models

Market Potential - Decomposition Methods III Market Potential - Decomposition Methods III


Choice Models - segmenting consumers and forecasting each segment

Choice Models - segmenting consumers and forecasting each segment

nmarket research basedä observed dataä hypothetical questions for new products/ services

Problems6 sample size in small populations6dynamics

ä how do the parameters change over the planning horizon6price dependency

ä measurement (quality)ä projected behaviour

Used in stable & growth markets


The Bass Model: basic segmentationThe Bass Model: basic segmentation

• Two types of people:– Innovators

• They adopt because of their attitude to technology

– Imitators • They adopt when exposed to

consumers who have adopted already

Red: adoptersBlack: potential adoptersBlue: lacking information

Red: adoptersBlack: potential adoptersBlue: lacking information

0

2

4

6

8

1 0

1 2

1 4

1 6

1 8

2 0

1 5 9 13 17 21 25 29 33 37 41 45 49 53 57

Logist ic

G o m p e r t z


Estimating the Diffusion Path:• limited if any sales data• S-Shaped curve used to represent adoptions• different curves and parametersØ different market potential and uptake trajectoryØ dependent on key parameters

Modelling multinational telecommunications demand with limited data, Islam et al, IJF, 2002

The Effect of Increasing Imitation in the Social System

0

5

10

15

20

25

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Hou

seho

lds

(Mill

ions

) Increases Slope of the Curve

The Effect of Increasing Innovation in the Social System

0

5

10

15

20

25

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Hou

seho

lds

(mill

ions

)

Pulls curve leftwards

Issues:• data limitations• market potential affected by product/ social factors


Models of Interacting IndividualsModels of Interacting Individuals

• Different networks of consumers Ø Different adoption patterns

Connected to your two nearest neighbours

Connected at RandomConnected to near neighbours but with other links

•Each group of individuals has its own rules of behaviour and interaction• affected by its environment

•Each group of individuals has its own rules of behaviour and interaction• affected by its environment

The impact of networked groups(Bass defines a simple network)


How can these ‘Agent based models’ be used (Collings, BTExact)

How can these ‘Agent based models’ be used (Collings, BTExact)

• Aim to understand behaviour of interacting markets– Diverse individual behaviour– Asymmetric information and motivation

• Examples – Financial markets– Customer relationship management


EXTENSIONS and ISSUES

• Can include marketing and economic variables

• Estimation (analogy or numerical methods)– Meta Models

• link the diffusion parameters to market characteristics• to other products, other markets

– Genetic algorithms

• Incorporation of effects from other products/ markets

Diffusion models

• Supply restrictions (in regulated markets)

• Disaggregate models ( e.g. to industry specific uptake of fax)


Bass Model - Diffusion Parameters and Meta Model

Imitation

Innovation

Mobile

NICs

UK, Denmark

Japan

Describing the speed and shape of the adoption path

Average diffusionparameters


Service provider

decision factors

influences

Potential Market

Marketing QualityPrice

Understanding Utility Acceptability

Adoption

Simulation Modelling• system dynamics replicate diffusion curves• offer more flexibility for incorporating

– information and processes– decision variables

Simulation Modelling• system dynamics replicate diffusion curves• offer more flexibility for incorporating

– information and processes– decision variables

Key benefit: transparency of model


Evaluation - Models of Market Potential

• Choice Models– Based on intentions data

• Models of Market Penetration– define potential = CM(t) where M(t) is

determined by the economic/ social system of the market

– Based on analogous products and countries

Problems • ‘current intentions• changing valuations

• unvalidated• c depends on time

Evaluation - Models of the Diffusion Path• Limited forecast validation

– short term (if any)– use too much data– poor benchmarks

• No forecast validation• Limited parameter validation

• Aggregate Bass-type diffusion

• Simulation


Final commentsFinal comments

• Widespread interest in telecoms forecasting

• Survey evidence suggests organisations which adopt a more ambitious and rigorous approach do better

• Primary methods ‘naïve qualitative’– despite major investments (and disasters) riding on the results of a forecas– Structured use of judgement (E.g. Delphi: Knut Blind)

• Too little academic research– too hard?– Too messy (Forecasting category sales and market share for wireless telephone

subscribers: a combined approach, Kumar et al, IJF, 2004)

IJF Seminar objectives:• identify where progress has been made• ‘gap’ analysis

IJF Seminar objectives:• identify where progress has been made• ‘gap’ analysis

Telecommunications demand – a review of forecasting · PDF fileTelecommunications demand...

Documents

Transcript of Telecommunications demand – a review of forecasting · PDF fileTelecommunications demand...